Question

我正在尝试将图片加载到网络并计算前5个错误率。我的数据层是

name: "CaffeNet"
layer {
  name: "data"
  type: "ImageData"
  top: "data"
  top: "label"
  image_data_param {
      source: "/home/filipa/PycharmProjects/Proposal_Code/Labels/val.txt"

  }
}

其中val.txt文件具有每个图像和对应标签的名称。（共50000张图片）

我的最后一层是

 name: "fc8"
  type: "InnerProduct"
  bottom: "fc7"
  top: "fc8"
  inner_product_param {
    num_output: 1000
  }
}
layer {
  name: "prob"
  type: "Softmax"
  bottom: "fc8"
  top: "prob"
}
layer {
  name: "accuracy_top_5"
  type: "Accuracy"
  bottom: "fc8"
  bottom: "label"
  top: "accuracy_top_5"
  accuracy_param {
    top_k: 5
  }
 include {
    phase: TEST
  }
}

我得到的错误是

E1116 11:34:28.173830  7465 io.cpp:80] Could not open or find file ILSVRC2012_val_00000001.JPEG
F1116 11:34:28.173835  7465 image_data_layer.cpp:72] Check failed: cv_img.data Could not load ILSVRC2012_val_00000001.JPEG

在我的代码中，我将图像加载到网络

im = caffe.io.load_image(IMAGE_FILE)

其中IMAGE_FILE指向我图像的确切位置。

Answer 1

我会将原型文件编辑为：

  image_data_param {
      root_folder: "/home/filipa/PycharmProjects/root/folder/for/data"
      source: "/home/filipa/PycharmProjects/Proposal_Code/Labels/val.txt"
  }

然后在＆＃34; val.txt＆＃34;中设置路径相对于root_folder。

如何使用Python将图像和标签加载到image_data_layer？

1 个答案: